Shopping Assistance for Everyone: Dynamic Query Generation On a Semantic Digital Twin As a Basis for Autonomous Shopping Assistance

Michaela Kümpel (University of Bremen), Jonas Dech (University of Bremen), Alina Hawkin (University of Bremen), Michael Beetz (University of Bremen)

Abstract

While the Digital Twin technology can be used by robotic agents to autonomously digitise retail stores, the Semantic Web offers vast machine-understandable product information that can be utilised by both digital and robotic agents. We propose connecting shopping assistants to a semantic Digital Twin for a service-oriented shopping experience. The semantic Digital Twin connects product information from the Semantic Web to retail environment information created by an autonomous robot performing stocktaking. It can be used to retrieve relevant information for action execution by shopping assistants that dynamically generate queries to answer complex questions like "Where is toothpaste from (my preferred brand) containing natural ingredients?", thus making the contained knowledge actionable.